Rational Adversaries and the Maintenance of Fragility: A Game-Theoretic Theory of Rational Stagnation

2510.22232v1 cs.GT, cs.AI, econ.TH 2025-10-29
Авторы:

Daisuke Hirota

Abstract

Cooperative systems often remain in persistently suboptimal yet stable states. This paper explains such "rational stagnation" as an equilibrium sustained by a rational adversary whose utility follows the principle of potential loss, $u_{D} = U_{ideal} - U_{actual}$. Starting from the Prisoner's Dilemma, we show that the transformation $u_{i}' = a\,u_{i} + b\,u_{j}$ and the ratio of mutual recognition $w = b/a$ generate a fragile cooperation band $[w_{\min},\,w_{\max}]$ where both (C,C) and (D,D) are equilibria. Extending to a dynamic model with stochastic cooperative payoffs $R_{t}$ and intervention costs $(C_{c},\,C_{m})$, a Bellman-style analysis yields three strategic regimes: immediate destruction, rational stagnation, and intervention abandonment. The appendix further generalizes the utility to a reference-dependent nonlinear form and proves its stability under reference shifts, ensuring robustness of the framework. Applications to social-media algorithms and political trust illustrate how adversarial rationality can deliberately preserve fragility.

Ссылки и действия

Связанные статьи

Collective dynamics of strategic classification

## Контекст Classification алгоритмы на основе Искусственного Интеллекта (ИИ) сегодня используются в высококонтрастных ...

2025-08-15